- Posted by admin
- On March 3, 2015
- AIFM, AIFMD, Liquidity Risk, Monte Carlo Simulation, Property, Real estate, REITS
An analysis of AIFMD as applied to Real Estate Investment Trusts and Property Companies
The Alternative Investment Fund Management Directive (AIFMD) is a piece of European Union legislation aimed specifically at “non-standard” funds that wish to operate and market themselves within the Union. In direct response to the recent financial crisis there are specific requirements in the area of the risk management of a fund which are both new and potentially quite onerous for a fund. The AIFMD which captures property investment funds specifically states (Article 44, para 2):
”The qualitative and quantative risk limits for each AIF shall, at least, cover the following risks: (a) market risk (b) credit risk; (c) liquidity risks (d) counterparty risks; (e) operational risks.”
In addition to giving the scope (very broad) of the required risk management function, the legislation also specified the timeliness of the operation of the risk management function – Article 44, paragraph 1a: states that an AIFM must:
“(a) identify, measure, manage and monitor at any time the risks to which the AIFs under their management are or might be exposed;”
This is a serious burden on the operation of any fund, but in the information scarce world of property funds, identifying all of the relevant risks is difficult, but amassing the information necessary to give an estimate of the magnitude of the risks and their correlation is doubly so. However the legislation goes further than just requiring the estimation of the current risk profile, there is also the requirement to effectively forecast the risk profile – look carefully again at the words above “…..the risks to which the AIFs under their management are or might be exposed;”
Thus many property investment managers will have to deal with a whole set of new regulations and requirements that they may well have been aware of but may have heretofore largely ignored. For example Directive 2011/61/EU:
“requires the Commission to specify the liquidity management systems and procedures enabling the AIFM to monitor the liquidity risk of the AIF”.
Article 58 of the Level 2 text specifies that:
“AIFMs should be able to demonstrate to their competent authorities that appropriate and effective liquidity management policies and procedures are in place. That requires due consideration to be given to the nature of the AIF, including the type of underlying assets and the amount of liquidity risk to which the AIF is exposed, the scale and complexity of the AIF or the complexity of the process to liquidate or sell assets.”
So whilst many property managers would have been acutely aware that property is a highly illiquid asset class (even in good times) AIFMD requires that they monitor the liquidity of the underlying assets and that they can model investor redemption profiles. The governance of real estate companies will also be impacted under AIFMD. For example in the original Directive (2011/61 EU Article 15) it is stated that:
“AIFMs shall functionally and hierarchically separate the functions of risk management from the operating units, including from the functions of portfolio management”.
A literal interpretation of this is taken to mean that there must be an effective Chinese Wall in place between the front office and the risk management function. In other words a property investment manager now has to either build a separate risk management function in-house or outsource to a professional risk analytics company.
Property Valuation Techniques
The return from property can be decomposed into two primary sources: capital return and income return. The income side of the portfolio primarily deals with the lease details (covenant, terms etc) and general risks to the ability of the tenant to comply with the lease terms in terms of paying over rent on a monthly, quarterly or however agreed basis. It can be seen that the prospect for capital gains (or losses) on the property depends primarily on the general macro-economic environment and the prospects for the real interest rate level. Like most financial assets property is valued as a stream of income discounted at a certain rate (the correct discount rate to be applied is quite a subjective matter and small changes in the rate used can make a large difference to the valuation of a long-tail asset such as property).
There is undoubtedly overlap between both the prospects for income and capital gains (or losses). For example in a recession there is likely to be a greater chance of void periods and subsequent income loss. Likewise as the economy emerges from recession there is likely to be the prospect of attracting new tenants and stronger rental growth. There are several well known methods to value property e.g. Sales Comparison, Capitalization, and Replacement Cost Methods. As an analogy to equity valuation and its associated measures of risk (using the Capital Asset Pricing Model) it can be seen that market risk (capital return) is analogous to beta in the CAPM framework whilst the selection and management of tenants is akin to alpha (colloquially known as manager skill). Thus allocating capital to purchase property in higher risk areas will demand a lower discount rate (higher return) and will likely require more management time. This is similar to equity managers allocating to higher risk equity markets such as emerging markets. The investment manager may be rewarded for allocating assets to these risky markets versus the “home” market; however he/she will not be rewarded for merely generating the market return similar to a passive portfolio in those regions (or indeed the home market). It is the managers skill in actively managing a property portfolio through active tenant management, development etc that generates alpha.
To focus on the asset level firstly it can be seen that the risk in a single property can encompass risks such as build quality, planning issues etc. At the portfolio level the risk becomes one of concentration, liquidity, leverage, counterparty etc. The standard CAPM model uses volatility as a measure of the riskiness of a portfolio of assets. Most investment managers (and indeed many academics) know that volatility does not equal risk (property development is considered the most risky aspect of property management and yet may be the least traded and hence appear least volatile). However it is a useful place to start.
One of the major problems in assessing risk in a property portfolio (the same applies by and large to private equity) is that property is not traded very often. Unlike the equity markets property transactions are irregular and often hidden from view, often taking place privately without disclosure to property registers or databases. Thus there are oftentimes wide dispersion’s between transactional based valuations (sales comparison techniques) and valuation based techniques (capitalisation or discounted cash flow techniques). This lack of data points results in poor time series data for comparison purposes but more importantly makes calculation of Value at Risk (VaR) metrics using historical data (non-parametric) difficult if not impossible – more on this later.
From speaking with several property managers it is clear to us that many property managers conflate risk management with valuation. A property investment manager may decide buying properties (that are well built, well located and occupied by tenants with good covenants) is low risk and that therefore this is good risk management. The property manager may well have reduced his or her exposure to known risks (e.g. the quality and credit worthiness of the tenant – assuming the information to make that judgement is readily available). However risk management is not just assessing, monitoring and then managing known risks but also involves performing similar analysis on unknown risks. This could, for example, be the financial institution that lends money to the property manager going bust while perhaps, at the same time, there is a sudden large redemption from an institutional investor. Whilst this apparent coincidence may have seemed fanciful a decade ago the financial crisis in 2008 brought home the reality that when bad things happen they tend to happen at the same time. This then is the essence of risk management – standing back dispassionately and saying “what if” and carefully evaluating the downside. The management of risk is most certainly not designed to eliminate (possibly not even to minimise) risks rather it is about taking advantage of opportunities in a carefully controlled manner and being able to survive the inevitable downturns (the “black swans”).
Risk Management Techniques
As outlined above due to the idiosyncratic nature of property investment conventional risk management techniques cannot be performed exactly as they are used for hedge funds for example. A typical hedge fund may trade highly liquid instruments and be priced daily by the administrator. This will not be the case for a property company. Therefore different techniques must be used such as those typically used in the fund of hedge fund space.
Determining the financial risk of a property portfolio is non-trivial and as such any implemented risk model would be highly influenced by the input of the relevant fund managers. It is fully appreciated that the valuation of the portfolio is of principal concern to the relevant managers, however it is the analysis of potential losses and the correlation of those losses with other losses in the portfolio that need to be addressed. In other words it is a requirement of the market risk management process that the risk factors associated with the valuation of the property(ies) must be assessed, and their magnitude determined. Furthermore the variability of these risk factors and the correlations between them must also be assessed. Currently there are two models that may be used to evaluate the financial risk. The first approach relates primarily to the discounted forward value of the property(ies) and as such may be characterised using market specific property indices along with an appropriate level of specific idiosyncratic risk.
A second approach is to forecast rental yields using exogenous data such as long term interest rates, GDP etc., and use a value/yield model to evaluate the sensitivity of the property portfolio to stressed conditions. It should be noted that in both models macroeconomic variables, such as GDP etc. will also have a real impact on the overall risk profile.
It should be stressed at this point that it is not the purpose of risk management to attempt to forecast future values of a single property or a portfolio thereof. The purpose of risk management is to determine the likelihood of the correlation between, and impact of, extreme events in the risk factors on the overall value of a single property or a portfolio of properties.
Counterparty and credit risk are closely related in a property portfolio. Credit risk may be increased depending on the leverage of the fund and the stability of the leases. In general it is envisioned that Monte Carlo simulation is an appropriate methodology to determine, monitor and manage the various financial risks of a property fund. Whilst there are undoubtedly a number of assumptions that will need to be made their impact may be included in the production of statistically significant risk metrics in the required areas of market, counterparty and credit risk.
According to the legislation liquidity risk is determined as a combination of investor liquidity in the fund and the liquidity of the investments of the fund. It should be noted that the legislators added a separate chapter into the AIFMD to cover liquidity risk and were quite specific in outlining the requirements for compliance with the provisions on liquidity risk. In this chapter they specify that both market and investor liquidity must be assessed under both business as usual (BAU) and stressed conditions and to ensure that the resultant liquidity profile of the fund is consistent with the information that the investors have been given.
Whilst traditional financial risk management has not gained common currency in property related funds to date, the requirements of AIFMD, and the likely increase of regulatory oversight of all funds going forward mean that far greater emphasis will have to given to risk management than was hitherto the case. This may mean considerable disruption for both funds and fund managers; however a robust risk management process will be seen to become a necessity both as a means of mitigating regulatory risk, but also a way to manage effectively the complex risks to which property funds are exposed.